Study uses machine learning to predict whether a patient is intoxicated due to pesticide exposure

02 Mar 2026

In response to a growing human population, greater attempts to correspondingly increase agricultural production become necessary. To boost crop production, pesticides have been widely applied against pests and plant diseases to attain enhanced crop efficiency. Unfortunately, routine application of pesticides brings about adverse effects unintentionally not only on human health but to the environment as well. Acute pesticide intoxication can happen due to the incorrect administration of pesticides, usually a high dose is either inhaled, ingested, or comes into contact with the skin or eyes. Other detrimental and dangerous health consequences can be attributable to the prolonged use of pesticides including problems involving the reproductive system, fetal maldevelopment, increased risk of cancer, and asthma. Identifying pesticide intoxication is important for implementing early intervention efforts.

This study aims to predict whether the patient is intoxicated due to pesticide exposure. Machine learning algorithms were trained and tested on an anonymized and publicly available pesticide exposure dataset.  It also identified which methodological enhancements to the feature selection step and class imbalance handling method would yield the highest predictive capability for pesticide intoxication. Results showed random forest with the most robust and optimum predictive capability generating the highest Matthews correlation coefficient and F1-score in all model configurations.

The integration of these machine learning tools in decision support systems for clinical observation of patients suspected of pesticide intoxication can be useful for swift diagnosis. This is a significant contribution to the health of agricultural workers while ensuring agricultural productivity. Future research can use another dataset with more varied pesticide intoxication. 

Authors: Hannah Gabrielle D. Buizon (Department of Physical Sciences and Mathematics, University of the Philippines Manila), Odedjinn Caezar Y. Suba (Department of Physical Sciences and Mathematics, University of the Philippines Manila), Rencio Noel Q. Simangan (Department of Physical Sciences and Mathematics, University of the Philippines Manila), Ma. Sheila A. Magboo (Department of Physical Sciences and Mathematics, University of the Philippines Manila) and Vincent Peter C. Magboo (Department of Physical Sciences and Mathematics, University of the Philippines Manila)

Read the full paper: https://ieeexplore.ieee.org/document/10908794

Image by zefe wu from Pixabay

Study uses machine learning to predict whether a patient is intoxicated due to pesticide exposure

In response to a growing human population, greater attempts to correspondingly increase agricultural production become necessary. To boost crop production, pesticides have been widely applied against pests and plant diseases to attain enhanced crop efficiency. Unfortunately, routine application of pesticides brings about adverse effects unintentionally not only on human health but to the environment as well. Acute pesticide intoxication can happen due to the incorrect administration of pesticides, usually a high dose is either inhaled, ingested, or comes into contact with the skin or eyes. Other detrimental and dangerous health consequences can be attributable to the prolonged use of pesticides including problems involving the reproductive system, fetal maldevelopment, increased risk of cancer, and asthma. Identifying pesticide intoxication is important for implementing early intervention efforts.

This study aims to predict whether the patient is intoxicated due to pesticide exposure. Machine learning algorithms were trained and tested on an anonymized and publicly available pesticide exposure dataset.  It also identified which methodological enhancements to the feature selection step and class imbalance handling method would yield the highest predictive capability for pesticide intoxication. Results showed random forest with the most robust and optimum predictive capability generating the highest Matthews correlation coefficient and F1-score in all model configurations.

The integration of these machine learning tools in decision support systems for clinical observation of patients suspected of pesticide intoxication can be useful for swift diagnosis. This is a significant contribution to the health of agricultural workers while ensuring agricultural productivity. Future research can use another dataset with more varied pesticide intoxication. 

Authors: Hannah Gabrielle D. Buizon (Department of Physical Sciences and Mathematics, University of the Philippines Manila), Odedjinn Caezar Y. Suba (Department of Physical Sciences and Mathematics, University of the Philippines Manila), Rencio Noel Q. Simangan (Department of Physical Sciences and Mathematics, University of the Philippines Manila), Ma. Sheila A. Magboo (Department of Physical Sciences and Mathematics, University of the Philippines Manila) and Vincent Peter C. Magboo (Department of Physical Sciences and Mathematics, University of the Philippines Manila)

Read the full paper: https://ieeexplore.ieee.org/document/10908794

Image by zefe wu from Pixabay